Fees and the true cost of a trade

The first cost is the spread. Buying at the ask rather than the midpoint costs you roughly half the spread immediately, and it never appears on a statement — it shows up as having bought higher than the price you would have written down as the market’s. On a four-point spread that is two points, every time you cross.

The second is the venue’s explicit fee, which is the one people compare. It is charged per trade, per contract, or on profit, depending on the venue.

The third is the cost of moving money. A fixed charge on a deposit or withdrawal is a percentage that depends entirely on the amount: $2 on a $50 transfer is 4% and on a $2,000 transfer is 0.1%. For a small account moving funds often, this is frequently the largest of the three, and it is the one that never features in a fee comparison table.

Venues do not charge in the same units, which is why a single “fee percentage” cannot be compared across them. The three common shapes are: a percentage of what the order cost you, a percentage of the profit if the position wins, and a flat amount per contract regardless of outcome. Each behaves differently, and the differences are large.

A percentage of cost is charged whether you win or lose, because it applies to the order rather than the result. A percentage of profit charges nothing on a losing position, which sounds generous and is expensive on a winner. A per-contract charge is indifferent to price, which makes it cheap on a 10¢ contract and brutal on a 95¢ one — because the same charge is levied against a much smaller possible gain.

This is also why our own fee tooling stores the shape alongside the rate, and shows “not recorded” rather than zero where a venue’s schedule is unknown. Printing $0.00 for an unknown cost would rank that venue top of any table sorted by net profit, which is a factual claim we would not be able to support.

Take $100 committed at 68¢. Contracts are indivisible, so that buys 147 of them — 100 ÷ 0.68 is 147.06, and the fraction is not for sale — committing $99.96 and leaving 4¢ unspent. If it settles YES, the payout is $147.00 and the gross profit is $47.04.

Under a fee of 1% of cost: 1% of $99.96 is $1.00, so the net profit is $46.04. Under a fee of 10% of profit: 10% of $47.04 is $4.70, so the net is $42.34. Under a flat 1¢ per contract: 147 × $0.01 is $1.47, so the net is $45.57.

Three schedules that all sound modest, three different answers, and the ranking between them changes with the price you paid and with whether the position wins. The only way to compare venues honestly is to run your actual intended position through each schedule, which is what a fee calculator is for.

Per-contract charges have a failure mode worth seeing explicitly. Commit $100 at 97¢ and you buy 103 contracts for $99.91. If it settles YES the payout is $103.00 and the gross profit is $3.09 — a thin margin, which is what a 97¢ contract is.

A flat charge of 3¢ per contract on that position is 103 × $0.03 = $3.09, exactly the profit. The position is right, and you have made nothing. At 4¢ per contract the fee is $4.12 and the net is −$1.03: a position that won and still lost money.

This is not a hypothetical curiosity. Buying high-priced contracts to collect small, reliable margins is one of the first strategies people arrive at, and it is the one most sensitive to fee shape. A schedule that is irrelevant at 30¢ can consume the entire thesis at 95¢.

Suppose you have done the work and believe an outcome is 65% likely, while the de-vigged market price is 62%. That is a three-point edge, which sounds like a business.

Now count the costs. The market has a four-point spread, so crossing it costs about two points relative to the midpoint. A 1% fee on the 64¢ ask is 0.64¢, roughly another 0.6 of a point. Total cost around 2.6 points against a 3-point edge, leaving 0.4 points — and that is before any cost of moving money, and assuming your 65% is right.

That arithmetic is why serious participants care about the spread far more than about the headline fee, and why a strategy that needs to trade often needs a much larger edge than one that does not. It is also why our cross-platform comparison refuses to call anything an opportunity until the gap survives half of each side’s spread plus each side’s fee. A gap that does not survive its own costs is not an opportunity; it is a way of paying for the privilege of being right.